generalized-linear-models

A Scalable Empirical Bayes Approach to Variable Selection in Generalized Linear Models

Haim Bar, James Booth and Martin T. Wells / 2021-03-13

In the toolbox of most scientists over the past century, there have been few methods as powerful and as versatile as linear regression. The introduction of the generalized linear model (GLM) framework in the 1970’s extended the inferential and predictive capabilities to binary or count data. While…

compositional-data

Compositional scalar-on-function regression as a tool (not only) for geological data

Ivana Pavlů, Karel Hron / 2021-03-10

Compositional data are characterized by the fact that the relevant information is contained not necessarily in the absolute values but rather in the relative proportions between particular components. As an example, take household expenditures for different purposes (housing, groceries, travel etc.)…

machine-learning

Higher Order Targeted Maximum Likelihood Estimation

Zeyi Wang and Mark van der Laan / 2021-03-10

Summary We propose a higher order targeted maximum likelihood estimation (TMLE) that only relies on a sequentially and recursively defined set of data-adaptive fluctuations. Without the need to assume the often too stringent higher order pathwise differentiability, the method is practical for…

compositional-data

Give me an adequate correlation: assessing relationships in percentage (or proportional) data

Karel Hron, Peter Filzmoser / 2021-02-04

We assume that you are quite familiar with the following problem. Consider a data set where the information is expressed in percentages or proportions. An example are household expenditures, given as average amounts (in Euros) the households are spending on food, housing, transportation, etc. Since…

webinars

Recent Advances in COVID-19 modelling

2021-02-04

YoungStatS project of Young Statisticians Europe, FENStatS, proudly announces our first One World YoungStatS webinar. With four young scholars, we will discuss Recent Advances in the Modelling of COVID-19, presenting novel statistical models, interesting advancements and applications of mechanistic…

classification

Locally adaptive k-nearest neighbour classification

Thomas B. Berrett / 2021-01-31

Binary classification is one of the cornerstones of modern data science, but, until recently, our understanding of classical methods such as the k-nn algorithm was limited to settings where feature vectors were compactly supported. Based on a new analysis of this classifier, we propose a variant…